Reject unreasonable customer requirements for high quality

 

Problem description

"We have to release before 5.1." I believe that any Software Project Team has met similar requirements from customers. Even more depressing is that before 5.1, customers may make other requests. The company boss or project manager may say the same thing to programmers again. The result is that most of them cannot be released before 5.1, or a version of very poor quality is released. The customer is angry and has serious consequences. The customer may refuse to pay or even never cooperate with you again, because you have not submitted any useful information.

Understanding of the problem

A version that must be released before version 5.1 means that the customer wants to release a high-quality version before version 5.1, rather than an unavailable version. If the time is fixed, the function should be able to be reduced. In most cases, the requirements cannot be defined in very detail. Not all functions can be reduced, but core functions are required in any case. If we find that even the core functions are not completed, we should inform the customer as soon as possible so that the customer can find other alternatives as soon as possible. The last moment let the customer know that it is a terrible thing to commit.

If the software required by the customer is a car, the engine is the most important core function, and the chassis, tires, and so on are the necessary auxiliary functions. Without these cars, the engine cannot run.

The earliest cars had no sound, no leather seats, no air conditioners, and no navigation, but it was hard to say that the quality of those cars was poor.

Essentially meeting customer requirements

A mature team should learn how to "satisfy the customer's requirements to the highest level and satisfy the customer's" essential "requirements, not just his verbal and superficial requirements.

This will naturally lead to many situations that require discounts, and the time is always limited. How should we make decisions? My idea is: we need to turn off the function. If the time is not enough, we can not release a function, but as long as a function is to be released, we need to do it well. For example, if the paging time is enough, we can allow the customer to define the number of lines per page. If the time is not enough, we can write it to death first, which will save some time, but the quality is still good.

If you want to cut off a slightly larger feature, you must clearly communicate with the customer, clearly tell the customer that the time is not enough, and discuss with the customer what to cut.

Quality is not discounted. This will become the culture of the store. If some customers do not care about quality, we would rather give up those customers.

Iterative development

Iterative development does not frequently commit bad versions, but rather provides some useful features.

Quality must be ensured for any submission. If quality cannot be ensured, no submission is required. Because low-quality submission customers are useless, it is equivalent to not submitting. The worse impact is that customers will think that your level is so bad. He liked the advertisement "the best or nothing" of Mercedes ".

Peter once mentioned a fast release mode for developing Internet applications. Quickly develop core functions, get online as soon as possible, and get customer feedback. After a small part of the market verification is successful, it will be fully rolled out. That fast mode does not conflict with high quality requirements. If the features you quickly release are difficult to use, you cannot get positive feedback.

There are countless new things on the Internet every year, which makes people think the speed is very fast. Do not be confused by this illusion. The vast majority of new things will become an eye-catching cloud. Really good products are well thought out. We cannot trust luck.

Digress

Finally, what should I do if my boss has arranged an impossible task for you? First, if this is normal, you should consider changing jobs. It is impossible to work in such a company. You can't change your job in a short period of time. You should raise your worries. If the boss doesn't agree, you can just smile and work hard, but you can't finish it if you can't finish it. The actual failure may make the boss realize that his decision is wrong. Generally, the boss is not too stupid and should be able to learn from the failure, if the customer is unhappy, the profit of the company will not be guaranteed.

Maintain a good attitude, exercise well, and keep learning. You should do this in any case. If you encounter a bad company, you should do it more. If the level improves, you don't have to worry about finding a good job.

Other digression

Finally, I would like to emphasize that the significance of a company lies in serving customers. This is also true for a project. All actions must be carried out around "serving customers. If you cannot meet all the requirements of the customer, it is natural to try to meet the most important requirements of the customer. If I want to buy a car that requires fast speed but cannot burn too much gasoline, the car manufacturer cannot meet my requirements. I can leave for the second place, speed is the most important thing for me. It is good to be able to consume less oil, but I can also accept cars with high speed but high fuel consumption.